I was actually interested in being a games journalist for a while. Aside from being caught between my journalistic integrity and the advertising dollars, though, there were other factors that turned me off from it.
Reviewers have a limited time available to play games before their articles have to go up. They need to form an accurate, strong and justified opinions about games often after just a few hours of play time. This doesn't just apply to the latest forgettable shooter, too, but to hundred-hour-long role-playing and strategy games that take at least half of that to get a good sense of the quality and depth of the title. Very rarely do they actually get the chance play these games enough.
The second problem is very much related to the first, and that is that journalists, because they play games for fairly short periods of time, are going to be drawn in more than others by technicalities. Pretty graphics? Check! Good voice-acting? Check! Nice camera angles and controls? Check! The problem with this approach is that reviewing games becomes more about reviewing the presentation aspects alone without grading the gameplay; while some reviewers are mindful of this, many are not. I have said many times (not on this forum, but elsewhere) that Eurogamer, for example, rates games based almost entirely on presentation, and when it comes to either a 6/10 or a 9/10, that is true in almost every single case.
Now, take an unconventional game like S.T.A.L.K.E.R. It's a first-person shooter, but it's slow-paced, it has technical issues everywhere, it lacks decent voice-acting (most of the game doesn't even have any), the graphics are good in stills but the animation is stiff, there's almost no coherent storyline the first time through the game, and it's very, very easy to take the game at face value and rate it based on technical merit alone, which is what many reviewers did when it came out (they did the same with the sequel too, although more justified in that case due to the number of crash-causing bugs it had on release).
However, playing and reviewing the game in this way just doesn't work. S.T.A.L.K.E.R. is an open-world sandbox game emphasising emergent gameplay and light role-playing elements. The best parts of the game, though, may not be felt until much later on. Let me try to paint a picture: the Zone is a large, fairly open world. Your goal is to get to the centre of it, but as you get closer and closer, you get farther and farther from civilisation, from safety and security. The enemies get more and more dangerous as you go in, use different weapons, etc. You need to stock up on food, drink, anti-radiation medicine, health kits and bandages, and tons of ammunition for your weapons. Coupled with a big bulky suit of armour that's required to survive the radiation and enemies present, this means that you're left fairly burdened and won't be able to carry much extra stuff with you. Now, in the game, there also happen to be artifacts: highly irradiated objects which give the player special benefits for wearing them, but more importantly can be sold for a large amount of money. The more stuff you're carrying, the slower you walk and the faster you run out of stamina. I think you can see where this is going. The farther and farther you get from civilisation, the closer you get to your objective, the more valuable the rewards get, but the less of them you can bring back with you, and the more dangerous the enemies and the environment get.
Needless to say, everyone who plays the game for a decent amount of time will get to the point where they are out of health kits and bandages, have two or three bullets left for their weapons, are fully loaded with artifacts to sell for a killing, and emerge from an underground bunker after completing an objective. However, you also become incredibly aware of your own mortality, because all of a sudden it occurs to you: I have to get back now. See, more than anything, S.T.A.L.K.E.R. is really a game of resource management, of weighing risk and reward. It may be a shooter on the outside, but it's all about using its environment and gameplay mechanics to their fullest to do one thing: instill a sense of fear and vulnerability in the player.
All of that above may be just sound a little too silly for some people here, but it's true; unfortunately, it's also an experience which takes a minimum of maybe fifteen or twenty hours to arrive at, as not only does it require some story progression, but also the culmination of progress up to that point. Throw the player into that situation from the beginning and it's going to have less impact; no, it has to build up over time. S.T.A.L.K.E.R. is a phenomenal game for precisely the reason I've described above, but people are much more willing to rate it on technical factors, on immediate, instant gratification than on the cumulative experience. Maybe that's because the reviewers don't have enough time, or simply don't know how to get that sort of thing out of the games they play. That's why I don't want to be a reviewer: it would kill the games for me.
More to the point of the thread, however, that is also why I don't really take reviews, especially of highly controversial games like Too Human, all that seriously. Mid sevens? Chances are the game is much better than you are initially led to believe, especially when the developers are very passionate about it. You can't blame Dennis Dyack or Peter Molyneux for the game failing to live up to your expectations. They are incredibly intelligent, talented individuals that care deeply about what they do, but as a developer, things happen. You need to hit a deadline, you run out of money, your team isn't as experienced as it could be, the hardware isn't as capable as you thought, the ideas that sound great from a design perspective just don't work that well. Too Human is a classic case of a game where people rate it down on technicalities like strange control schemes, inconsistent presentation, bugs, etc.; in otherwords, all the stuff that their spokespeople (or effigies, if you prefer) are generally not responsible for. Admittedly, I have not played it; this is just the impression I get from the outside, simply because I have seen it in so many games. The fact of the matter is, what separates a good game from a great game these days has nothing to do with gameplay or design, it has everything to do with how pretty and immediately accessible your game is, and it's a crying shame. Reviewers, and consumers, have to realise that not all games can be perfectly polished blockbusters, even when they want to be, and to look a bit deeper at games to figure out what makes them really tick.
I also know that nobody here will read this post, because it's a bit too long for most of you kids. If you got this far, pat yourself on the back.